Model benchmarks

Qwen3.6-27B-MLX-8bit local LLM performance

As of July 2026, Qwen3.6-27B-MLX-8bit runs at up to 11.3 tok/s for local inference (best of 4 community benchmark runs across 1 GPU).

oMLX

Model size

27B

Peak speed

11.3 tok/s

Average speed

7.5 tok/s

Min memory

32.9 GB

Max context

6,687 tokens

Best quality

91.0

Benchmark runs

4

GPUs tested

1

Performance by hardware and tool

Every hardware/tool/quantization combination Qwen3.6-27B-MLX-8bit has been benchmarked on, ranked by peak token generation speed. Last updated July 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
Apple M5 MaxoMLX11.3 tok/s7.5 tok/s32.9 GB6,687 tokens76.64

Frequently asked questions

How fast is Qwen3.6-27B-MLX-8bit for local inference?
Across 4 community benchmark runs, Qwen3.6-27B-MLX-8bit reaches up to 11.3 tok/s and averages 7.5 tok/s, with the fastest results on Apple M5 Max.
How much memory does Qwen3.6-27B-MLX-8bit need?
The leanest observed configuration used about 32.9 GB of memory.
Which tools have been used to run Qwen3.6-27B-MLX-8bit?
Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.